An analysis of constrained robust regression estimators

Multicollinearity and nonnormal errors are problems often encountered in the application of linear regression. Estimators are proposed for dealing with the simultaneous occurrence of both multicollinearity and nonnormality. These estimators are developed by combining biased estimation techniques with certain robust criteria. An iteratively reweighted least-squares procedure is used to compute the estimates. The performance of the combined estimators is studied empirically through Monte Carlo experiments structured according to factorial designs. With respect to a mean-squared-error criterion, the combined estimators are superior to ordinary least-squares, pure biased estimators, and pure robust estimators when multicollinearity and nonnormality are present. The loss in efficiency for the combined estimators relative to least squares is small when these problems do not occur. Some guidelines for the use of these combined estimators are given.     

The capital supply curve in capacity expansion models: Some economic and algorithmic aspects

Capacity expansion models typically minimize the discounted cost of acquisition and operation over a given planning horizon. In this article we generalize this idea to one in which a capital supply curve replaces the usual discount rate. A capital supply curve is a means to model financial outlook, investment limits, and risk. We show that when such a curve is included in a capacity expansion model, it will, under certain conditions, provide a less capital intensive solution than one which incorporates a discount rate. In this article, we also provide an algorithm that solves capacity expansion models that incorporate a capital supply curve. The attractive feature of this algorithm is that it provides a means to utilize the “discount rate” models efficiently. Throughout, we give applications in power generation planning and computational experience for this application is also presented.     

Inequalities involving the lifetime of series and parallel systems

Let {Xⁱ} be independent HNBUE (Harmonic New Better Than Used in Expectation) random variables and let {Yⁱ} be independent exponential random variables such that E{Xⁱ}=E{Yⁱ} It is shown that \documentclass{article}\pagestyle{empty}\begin{document}$ E\left[{u\left({\mathop {\min \,X_i}\limits_{l \le i \le n}} \right)} \right] \ge E\left[{u\left({\mathop {\min \,Y_i}\limits_{l \le i \le n}} \right)} \right] $\end{document} for all increasing and concave u. This generalizes a result of Kubat. When comparing two series systems with components of equal cost, one with lifetimes {Xⁱ} and the other with lifetimes {Yⁱ}, it is shown that a risk-averse decision-maker will prefer the HNBUE system. Similar results are obtained for parallel systems.     

Optimal location of public facilities

Location of both public and private facilities has become an important consideration in today's society. Progress in solution of location problems has been impeded by difficulty of the fixed charge problem and the lack of an efficient algorithm for large problems. In this paper a method is developed for solving large-scale public location problems. An implicit enumeration scheme with an imbedded transportation algorithm forms the basis of the solution technique.     

Minimum-cost flows in convex-cost networks

An algorithm is given for solving minimum-cost flow problems where the shipping cost over an arc is a convex function of the number of units shipped along that arc. This provides a unified way of looking at many seemingly unrelated problems in different areas. In particular, it is shown how problems associated with electrical networks, with increasing the capacity of a network under a fixed budget, with Laplace equations, and with the Max-Flow Min-Cut Theorem may all be formulated into minimum-cost flow problems in convex-cost networks.     

一种电路模块故障诊断的实现方法

在分析电路模块故障诊断特性的基础上,本文利用表结构表示推理规则,以框架结构组织规则库,为快速孤立故障,将症状按其检测的难易程度划分类别,并由检测点的选择准则确定下一个能有效孤立故障的检测点,该方法在“电子装备模块级故障诊断系统”和某装备测距系统与搜索系统的实际故障处理中得到了应用,并证实了其有效性。     

STD-BITbus分布式测试系统

由STD总线和BITbus串行通信总线构成的分布式自动测控系统,具有实时性强、可靠性高、成本低等特点,可广泛应用于武器系统的综合测试和各类仓库的监控管理,本文介绍了STD-BITbus分布式测试系统的系统构成、软件框图和模拟实验结果。     

针对复杂检测系统的微机数据管理系统设计方法探讨

本文以陆军战术导弹性能检测系统为研究对象,探讨复杂检测系统中微机数据管理系统的设计方法,给出了此类系统标准化设计的规范步骤。     

Turbo C与Microsoft FORTRAN混合编程的实现

本文通过研究Borland公司的Turbo C与Microsoft公司的FORTRAN混合编程时出现的问题,提出了修改初始化模块及函数库的原理与方法,并利用TLINK的动态连接功能,实现了这两种语言在源程序级的混合编程,最后说明了实现过程与注意事项。     

一种基于中值滤波和边缘检测中模板匹配的图象滤波增强算法

本文针对数字图象处理中滤波可使圈象边缘模糊的现象,设计了一种组合式滤波增强算法。该算法在边缘地区沿着边缘方向进行滤波增强,在其它地区则进行滤波平滑操作,消除噪声。